Cargando…
Development of a hardware-accelerated simulation kernel for ultra-high vacuum with Nvidia RTX GPUs
Molflow+ is a Monte Carlo (MC) simulation software for ultra-high vacuum, mainly used to simulate pressure in particle accelerators. In this article, we present and discuss the design choices arising in a new implementation of its ray-tracing–based simulation unit for Nvidia RTX Graphics Processing...
Autores principales: | , , , , |
---|---|
Lenguaje: | eng |
Publicado: |
2021
|
Materias: | |
Acceso en línea: | https://dx.doi.org/10.1177/10943420211056654 http://cds.cern.ch/record/2807904 |
_version_ | 1780973074242338816 |
---|---|
author | Bähr, Pascal R Lang, Bruno Ueberholz, Peer Ady, Marton Kersevan, Roberto |
author_facet | Bähr, Pascal R Lang, Bruno Ueberholz, Peer Ady, Marton Kersevan, Roberto |
author_sort | Bähr, Pascal R |
collection | CERN |
description | Molflow+ is a Monte Carlo (MC) simulation software for ultra-high vacuum, mainly used to simulate pressure in particle accelerators. In this article, we present and discuss the design choices arising in a new implementation of its ray-tracing–based simulation unit for Nvidia RTX Graphics Processing Units (GPUs). The GPU simulation kernel was designed with Nvidia’s OptiX 7 API to make use of modern hardware-accelerated ray-tracing units, found in recent RTX series GPUs based on the Turing and Ampere architectures. Even with the challenges posed by switching to 32 bit computations, our kernel runs much faster than on comparable CPUs at the expense of a marginal drop in calculation precision. |
id | cern-2807904 |
institution | Organización Europea para la Investigación Nuclear |
language | eng |
publishDate | 2021 |
record_format | invenio |
spelling | cern-28079042022-04-29T19:17:00Zdoi:10.1177/10943420211056654http://cds.cern.ch/record/2807904engBähr, Pascal RLang, BrunoUeberholz, PeerAdy, MartonKersevan, RobertoDevelopment of a hardware-accelerated simulation kernel for ultra-high vacuum with Nvidia RTX GPUsDetectors and Experimental TechniquesMolflow+ is a Monte Carlo (MC) simulation software for ultra-high vacuum, mainly used to simulate pressure in particle accelerators. In this article, we present and discuss the design choices arising in a new implementation of its ray-tracing–based simulation unit for Nvidia RTX Graphics Processing Units (GPUs). The GPU simulation kernel was designed with Nvidia’s OptiX 7 API to make use of modern hardware-accelerated ray-tracing units, found in recent RTX series GPUs based on the Turing and Ampere architectures. Even with the challenges posed by switching to 32 bit computations, our kernel runs much faster than on comparable CPUs at the expense of a marginal drop in calculation precision.oai:cds.cern.ch:28079042021 |
spellingShingle | Detectors and Experimental Techniques Bähr, Pascal R Lang, Bruno Ueberholz, Peer Ady, Marton Kersevan, Roberto Development of a hardware-accelerated simulation kernel for ultra-high vacuum with Nvidia RTX GPUs |
title | Development of a hardware-accelerated simulation kernel for ultra-high vacuum with Nvidia RTX GPUs |
title_full | Development of a hardware-accelerated simulation kernel for ultra-high vacuum with Nvidia RTX GPUs |
title_fullStr | Development of a hardware-accelerated simulation kernel for ultra-high vacuum with Nvidia RTX GPUs |
title_full_unstemmed | Development of a hardware-accelerated simulation kernel for ultra-high vacuum with Nvidia RTX GPUs |
title_short | Development of a hardware-accelerated simulation kernel for ultra-high vacuum with Nvidia RTX GPUs |
title_sort | development of a hardware-accelerated simulation kernel for ultra-high vacuum with nvidia rtx gpus |
topic | Detectors and Experimental Techniques |
url | https://dx.doi.org/10.1177/10943420211056654 http://cds.cern.ch/record/2807904 |
work_keys_str_mv | AT bahrpascalr developmentofahardwareacceleratedsimulationkernelforultrahighvacuumwithnvidiartxgpus AT langbruno developmentofahardwareacceleratedsimulationkernelforultrahighvacuumwithnvidiartxgpus AT ueberholzpeer developmentofahardwareacceleratedsimulationkernelforultrahighvacuumwithnvidiartxgpus AT adymarton developmentofahardwareacceleratedsimulationkernelforultrahighvacuumwithnvidiartxgpus AT kersevanroberto developmentofahardwareacceleratedsimulationkernelforultrahighvacuumwithnvidiartxgpus |